"You are grounded!": Latent Name Artifacts in Pre-trained Language Models.
Vered ShwartzRachel RudingerOyvind TafjordPublished in: EMNLP (1) (2020)
Keyphrases
- language model
- pre trained
- language modeling
- training data
- n gram
- document retrieval
- training examples
- query expansion
- probabilistic model
- test collection
- speech recognition
- information retrieval
- retrieval model
- language modelling
- control signals
- statistical language models
- document ranking
- relevance model
- data sets
- language models for information retrieval
- multi modal
- training samples
- small number
- knn
- neural network